Peter Turcan

Trireme Commander

Computer simulation of Greek and Persian triremes

Like many people, I was very impressed with the Anglo-Greek reconstruction of the trireme "Olympias”, completed in 1987. Historical research at its most revealing - build the ship and test it for real. This is the only sure way of being right about the capabilities of ancient inventions - whether as relatively simple as a catapult or ballista, or as complex as a rowing galley. The building and sea trials of Olympias gave us a lot of technical data on the speeds, acceleration, turning circles, stopping distances and so on, of an Athenian trireme. The results are reliable because of the accuracy of the plans the Athenians left behind, and of course the careful reconstruction.

However, even with a lot of good data there are holes - performance in wind, going backwards, performance when damaged, and of course, how well the vessel did when ramming another, or being rammed.


Trireme battle video

An Athenian trireme has three levels of rowers, each with a single oar, with 85 rowers on each side of the ship. The main weapon is its heavy bronze ram - used for both ramming and for raking (shearing another vessels oars off). Secondary weapons include the arrows and javelins from the archers and hoplites on deck.

The simulation of an individual trireme by computer is simply the first step into simulating trireme battles - so on top of an accurate simulation
of the ships themselves, there would have to be an AI (artificial intelligence) system that would enable realistic decision making by a multitude
of ship's captains, and, at a higher level, squadron and fleet commanders. The AI would have to include a signaling system between ships that matched what we know of ancient navies, and a very robust attempt at modeling the formations that fleets took to navigate from one place to the next. In particular, moving in line astern, the only safe formation for any fleet to take, ancient or modern. Robust in this context means navigate a line of triremes in and out of port, around islands, through narrow straits, from line astern to line abreast, avoiding each other, and so on.

Given that the trireme plans were Athenian, it made most sense to focus on the battle of Salamis - the defeat of the Persian navy lured into the Straits of Salamis, by a smaller but still sizable Athenian fleet. However, the battle of Salamis would be the last scenario in a series, starting from the very beginning - starting, stopping, and maneuvering a trireme.

Working on and off on this, and taking too many years over it, I finally came up with Trireme Commander, a game of sorts but mostly a simulation - there are no features that enhance the capabilities of the triremes to make the simulation more exciting, fast, explosive, fiery or whatever. Playing the game should educate the user to the point where they understand clearly the mechanics of triremes and the tactics and pace of the battles.

Given there is a lot to learn, I decided on giving the player’s ranks, and as the player completed various tutorials and challenges, promote them so that the scenarios get progressively more involved - with more ships and more complexity.

The first scenario, and the only one for newcomers to the game, is called the Helmsman Tutorial. In this, the player is first given a presentation (OK, a bit of a lecture!) on how to control a trireme, then given an obstacle course to navigate.

Helming a trireme under oars is an interesting experience. Like any rowing vessel, the rowers face the stern. The 85 rowers on one side can only do the same movement, however the left and right sides can do different movements. There are basically four options: rowing (the default, pulling on the oars), backing (pushing rather than pulling the oars so the trireme goes slowly in reverse), dragging (dragging the oars in the water to have a braking effect) and lifted (simply holding the oars out of the water).

Normally, both sides row, and a stroke rate sets the acceleration - top speeds are around 8 or 9 knots (one knot is about 2000 yards, bit over a mile). If one side does a different movement to the other, there is obviously a turning effect. To steer the trireme, there are two heavy wooden rudders. These rudders create a lot of drag, and a decent amount of steering can come from having just one half of one rudder immersed, and the other rudder completely out the water. Putting more rudder into the water adds to the turning effect, but slows the trireme down.

The “emergency stop” of a trireme is to drag both sets of oars, and fully immerse both rudders. Even then it takes a while to stop the trireme, making collisions among friendly vessels a constant hazard.

Another advantage of having two rudders of course is that if one gets sheared off in battle, you still have the other one. Without any rudders some steerage comes from the rowers themselves, having one side lift the oars whilst the other rows for example, though this is cumbersome and only used as a necessity in retreat. Sails were not used, and often not carried, in battle. Booms were left in place though to display signal flags.
Mastering the basic controls is the purpose of the first tutorial. Bring the trireme, named Pandora, back to the dock and you pass - promotion from the insulting title of Landlubber to the lofty rank of Cadet.

As well as leaving behind plans for a trireme, the Athenians left behind the names of 100s of their vessels - with wonderful names such as Pandora, Olympias, Phosphoros, Aphrodite and many more.

As a cadet, it is time for some action. There are two cadet scenarios to pass. The first involves racing some other triremes through the training area (the obstacle course of planks and piers used for the initial tutorial). It is fun watching the AI controlled triremes choose their rowing styles and rudder movements to navigate the course. In this race, it is not necessary to come first to pass, simply to do a good time. You do not pass if your trireme, or any of the others, are damaged!

Next, having mastered the movement of a single trireme, it is time for a different kind of test - ramming an enemy vessel. In the Ramming trials scenario this is made easy by having a captured and empty enemy trireme left afloat - and your task is simply to smash into it. The mechanics of a ram are not too complicated - the more momentum at an angle as close to 90 degrees as possible will cause the most damage.

The target in this case is a Persian trireme, but much less is known about these boats as plans have not been left behind. There is some indication in ancient accounts that these triremes were lighter than the Athenian triremes. I decided to reduce the rowers on each side by 9, so there are 76 rowers on each side of a Persian trireme, and reduce the displacement by 4,000 Kg. I don’t have any figures to back this up, but these numbers slightly reduce the effectiveness of the triremes, but still leaves them very capable. These two numbers are the only differences in Trireme Commander that affect performance between Athenian and Persian triremes, I did not dumb down the Persian triremes, AI or weaponry in any other way.

Following success in the race, and the ramming trials, you are promoted to Helmsman (progress, but still a long way to go to Admiral). The Helmsman challenges start to get more warlike. The first is to sink a lonesome Persian trireme that is damaged and struggling to get back to its port (the Easy Target scenario). Next, there is a single Persian trireme, but it is not damaged - the One on One scenario. In these simple battles you will see the effectiveness of ramming, and that in fact most vessels capsize rather than sink as a result of ramming - they are after all made of wood and do not have particularly high swamp angles (the angle at which the boat will not rock back upright, but instead take on water and keep going over).

Perhaps at this stage the brutality of trireme battles becomes apparent. The trauma of being trapped in a capsizing boat being followed by a perilous swim to a dangerous shore, and that’s the lot of the lucky ones who escape. You will also see the first use of the archers and hoplites on board. They have nowhere near enough destructive power to disable a trireme, but nevertheless cause harassment, and injuries in small numbers.

Trireme battles feature very occasionally in Hollywood. The movie that sticks in my mind is the 1959 Ben Hur - which got several things right (the tedious remake had the aftermath of the battle with ships upright and on fire, duh). The original Ben Hur had the rowers as slaves (though many were actually free men and professionals, to a degree), but did have the famous “ramming speed” scene of increasing stroke rates, had ships raking (shearing others oars off), had a lot of fear and a great degree of confusion - all realistic. Something to compare this simulation with. Modern movies of ancient battles - land and sea - have way too much fire and effective missile throwing, for a visual effect that gets a zero for historical accuracy.

The next two scenarios for the Helmsman do not involve any fighting. The first is a tour of the Straits of Salamis, to familiarize yourself with the Straits - the terrain, islands and narrow passages. In this scenario though you have a small squadron, three ships are following you in line astern. Maybe there are some Persian ships around, but your orders are not to engage. Get back to your assigned port with no damage.

The final Helmsman scenario is a tutorial on the signaling system. We do know that signaling was handled by flags - along with shouts, loud instruments, and the like, but have few details. To be realistic the signaling cannot be too complicated (an order like “go and ram that vessel 45 degrees off your port bow” was probably impossible). The signaling takes the form of choosing a battle formation, with some additional signals such as indicating that you are repairing your vessel, or require assistance. I did like some of the older pictures of galleys, showing flags as hanging triangles. The big advantage of hanging a flag this way (over a flag pulled up a pole) is that the flag displays fully without any wind.

All that is required of passing this tutorial is to take a small fleet (3 squadrons, 12 triremes in total) out to sea, line up in a battle formation, dissolve the formation, and return safely. Light wind. No enemy in sight.

The battle formations are the Periplus (a flank attack in line astern), the Diekplus (a line-abreast frontal charge, aimed at raking the enemy to leave them oarless and floundering before ramming - sometimes by a second or third line), and the intriguing and perhaps doubtful Kyklos - a defensive circle.

Passing all four challenges will get you promoted to Captain (or “Trierarch” to the ancient Greeks). Now you have some real clout.

There are three Captain’s challenges, in different locations around the Island of Salamis. In each you command a squadron - so you have subordinates who will follow your signals to the letter (mostly). The first scenario gives you a slight numerical advantage (the Upper Hand scenario), the next you have an equal number of triremes as the Persians (the Even Odds scenario), and finally a tough one (Against the Odds) where you have eight triremes to the Persian’s ten. Even with the slightly heavier Athenian trireme this is a tough scenario to win. You don’t need crushing victories to pass these challenges, just come out on top.

You will notice that sometimes it is difficult to clearly tell if a trireme ahead of you is friendly or enemy - despite the different flags, booms and sizes of them. This fits with the historical accounts, which have Artemisia (queen of Halicarnassus, a Greek state but fighting for the Persians at Salamis) deliberately ram a friendly vessel to escape the attention of an approaching Athenian trireme. If a veteran captain can be fooled this way, then clearly identification can be very difficult. Artemisia, incidentally, is one of the very few female naval commanders from history - was clearly ruthless - and you can imagine the treatment Hollywood and fantasy history gives her.

In the game there is the difficult decision of when a scenario actually ends. Originally I had it when you returned to port, but as this can take several hours of hard rowing (and as this is a realistic simulation this translates to several hours of hard simulating), I changed this to enable an early evaluation when the flagships had parted a sufficient distance.

If you can win these three scenarios you are making serious progress - to the rank of Squadron Commander. There are just two scenarios at this level (Squadrons Skirmish and Windy Chase), where you now have three squadrons - one following you and two others - to signal. The additional squadrons allow for the greater tactical flexibility of wings. You can signal one wing (on your left or right) to engage and leave the center and other wing back. Or signal both wings to engage, or signal one wing to heave-to as you engage. Or any other combination.

Windy Chase is the only scenario featuring stronger winds. Wind has the effect of turning a ship side on to the wind’s direction (“broaching” in sailing lingo), and makes steering in a straight line difficult. It is not difficult though to see that once wind reaches a certain strength all battles are off, surviving and escaping the wind becomes the sole focus.

In these larger scenarios you will see that it can take a while simply to form up a fleet - always from a line astern formation - into a battle formation such as a two or three line Diekplus. All the triremes are controlled by an AI captain and they have to steer their vessels into the correct position, or close enough, just as you do. When they are in the right spot, they will hoist an “in position” flag. One of the crucial decisions of a commander is when to engage. Do you wait for all triremes to be in position, or most, or just a few?

The Commander (and Admiral) scenarios break from the Athenian-side-only perspective, and can be played from either the Athenian or Persian flagships.

The only gesture I have made in the simulation towards land forces is the existence of defensive forts around the ports. If truth be told, most triremes were simply pulled up onto a beach and did not have a dock to go to, and, whereas the existence of catapults is certainly known, their use in defending ports is a stretch. Their claimed range of 500 yards though makes this role at least plausible. Suffice to say if you get too close to an enemy port expect to have whistling catapult bolts drop from the sky.

Finally, promotion to Admiral. At this level not only do you have multiple squadrons, but there are other fleets on your side, but not under your command, involved in the battles. A degree of chaos is inevitable.

Try holding, or breaking open, the Kyklos - the famous circle of ships the Athenians used at the battle of Artemisium (fought a few weeks before Salamis). Try fighting some fictional scenarios that came close to reality: a battle in the open sea and not in the straits (giving the Persians room to form up their massive fleet), and a battle between the Egyptian fleet (Persian allies) and the Corinthian fleet (Athenian allies) on the opposite (West) side of the Island of Salamis. These are big events, and by now you will be familiar with the aftermath of a trireme battle: no fires, no smoke columns, but capsized hulls, sunken vessels with masts still sticking out the shallow water, and stricken and run-aground triremes that have little hope of repair or escape.

And of course there is the Battle of Salamis itself. Choose your role, your strategy, your targets, and try to make sense of the immense confusion over what is really going on.

YouTube videos
Trireme squadron battle

This video shows a smaller trireme battle, just one squadron on each side.
Form up, approach, and attack
This video shows a large force of triremes form into an attacking formation, then approach and engage the enemy fleet.

For some insights into the decision making of the computer controlled captains, refer to Artificial Intelligence .

Technical notes

Trireme Commander is a game for a Windows PC.
It requires that the Microsoft XNA Framework Redistribuatble 4.0 be installed first (an easy install from here ).
Runs on Windows 7/8/10. Onboard graphics (1600 x 900 minimum resolution) will work, a graphics card is preferrable for the biggest battles.
Audio is important, external speakers add a lot. Input is keyboard and mouse, and it is single player only.
Send me a message.
Peter Turcan Ph.D, a racing kayaker, military historian and computer programmer.

Read the interview with
Copyright © 2017  Peter Turcan. All Rights Reserved